CN102422144B - Automatic analysis device - Google Patents

Abstract

In order to shorten the measurement time of biochemical measurement, an indicator of the completion of the reaction is required in order to change the photometry time for each item or change the measurement time for each sample, but there is no method for judging the completion of the reaction until now. In the measurement of the measurement target substance contained in the sample, the parameters of the approximate formula are calculated using the measured values that change over time, and the degree of convergence of the reaction is judged based on the degree of convergence of the parameters, and the degree of convergence of the reaction is judged using The parameters of the time point are used to calculate the measured value of the reaction completion time point.

Description

Automatic analysis device

technical field

The present invention belongs to the technical field of automatic analysis devices for qualitative and quantitative analysis of multiple items on biological samples such as blood and urine, and particularly relates to a device capable of monitoring various biological samples contained in biological samples over time. An automatic analyzer with the function of measuring the degree of influence of components on the target substance to be measured and performing the measurement.

Background technique

An automatic analyzer for clinical examination dispenses a certain amount of sample and reagent, stirs it, and reacts it. By measuring the absorbance of the reaction solution for a certain period of time, the concentration of the component to be measured is obtained based on the measurement result.

The number of tests per hour is used as an indicator of the processing capacity of the device. Since the automatic analyzer was developed, in addition to the improvement in the accuracy of the measurement results, the improvement in the processing speed of the device has been gradually developed by many manufacturers of automatic analyzers. In order to improve the processing capacity of the device, increase the number of reaction cells that can be used (enlarge the size of the device), or increase the dispensing speed of the sample and reagent probe (speed up the probe movement), or improve the transport line of the sample rack High-speed and efficient, high-speed data processing capabilities of PCs, etc. As a result, the time from blood collection to reporting of measurement results is significantly shortened. One of the limiting factors for the measurement processing speed of these current high-throughput automatic analyzers is the reaction time between the sample and the reagent at the time of measurement, which depends on the reactivity of the reagent. The reaction time of the biochemical analysis device is usually about 10 minutes for each item. Depending on the project, the completion time of the reaction between the sample and the reagent is different, and the measurement method of clinical examination can be divided into two types: the endpoint method and the rate method according to the analysis method.

In the endpoint method, the change in absorbance decreases with time and eventually asymptotically reaches a constant value (final absorbance). The concentration of the component to be measured in the sample is obtained from the asymptotic absorbance value. In the endpoint method, there are items that reach the final absorbance at an earlier stage, such as T-CHO (total cholesterol) and Glu (glucose), as well as CRP such as CRE (creatinine), TP (total protein), and immunoturbidimetric method. (C-reactive protein), IgA (immunoglobulin A), IgG (immunoglobulin G), IgM (immunoglobulin M), etc. react slowly, become the final stable state, and take time to reach the final absorbance reaction.

The rate method is an inspection method for usually measuring the progress rate of a reaction starting from the reaction between a sample and a reagent. In the rate method, the speed of absorbance change is roughly constant, and the reaction process is a straight line. In the rate method, enzymatic method, etc., the reaction continues until the substrate or coenzyme is consumed, so the absorbance will continue to rise or fall, and will not reach a constant value, except for high cases where the sample concentration exceeds the allowable range. Therefore, the activity value of the item is calculated not from the concentration of the enzyme itself but from the rate of change of the absorbance of this straight line. However, if the reaction stops and the absorbance changes rapidly within the measurement time used for velocity calculation, the concentration of the item cannot be accurately measured if the absorbance at that point is used, so the reaction velocity is calculated without using the absorbance at this photometry point. In Patent Document 1, a calculation method that eliminates the effort of dilution and re-inspection, etc. may be employed.

When a sufficiently long reaction time cannot be obtained, as a method of obtaining a good measurement result, for example, Patent Document 2 discloses using the measured time and absorbance data, and using y=A+(B-A )/exp(Kt) approximates the relationship between absorbance and time. Among them, A is the final absorbance, B is the initial absorbance of the reaction, K is the reaction rate constant, and t is the measurement time. In this method, the concentration of the substance to be measured is obtained based on the obtained A, B, and K.

prior art literature

patent documents

Patent Document 1: Japanese Patent Application Laid-Open No. 1-59041

Patent Document 2: Japanese Patent Application Laid-Open No. 6-194313

Contents of the invention

The problem to be solved by the invention

In the field of clinical examination in hospitals, it is required to report the measurement results of patient samples as quickly as possible. Especially in emergency medical treatment at night, accident scene, diagnosis and treatment scene and other places that need emergency treatment, it is necessary to get the result as quickly as possible. Recently, there are also hospitals that conduct pre-examination examinations for general patients, that is, measure patient samples before examinations, and conduct examinations and treatments based on the examination results during examinations. Since the test results are known at the time of the initial consultation, the burden of patients having to visit the hospital again to inquire about the results can also be reduced. Of course, the time from blood collection to report of test results is also desired to be short for the patient and the diagnosis and treatment side. In clinical examination, when analyzing the time from blood collection to test result report, it is divided into the following three categories: 1) The storage time from blood collection to centrifugation; 2) The time from placing the sample in the analysis device until the completion of the measurement; 3) In addition, time for sample delivery, centrifugation, registration of patient information, reporting of results, etc. 3) The sample delivery and data processing time can be greatly reduced by the sample delivery system and the systemization of the entire examination room. The storage time of 1) has also been shortened by the development and popularization of high-speed coagulation type blood collection tubes. On the other hand, regarding the reaction time in the measurement of the automatic analyzer of 2), the reaction time of 10 minutes has not changed during these 30 years. In current automatic biochemical analyzers, etc., a system is incorporated in which emergency samples are inserted between routinely performed general samples and measurements are performed preferentially. However, the measurement time of the reaction between the sample and the reagent is the same as that of a general sample, that is, it takes a fixed time of about 10 minutes at least until the result is obtained after placing the sample. As the speed is increased, it is also necessary to shorten the reaction time in order to further speed up the measurement. However, if the measurement time is only shortened, accurate results cannot be obtained by measuring the concentration and activity value from the absorbance at the point in time when the reaction is incomplete.

As reagents for component measurement of biological samples, enzyme reactions, antigen-antibody reactions, chelation reactions, electrode methods, and the like are used. In the electrode method for measuring K (potassium) and Na (sodium) ions in the sample, the time until the measurement is completed is as short as about 1 minute. In the chelation reaction used in the measurement of inorganic substances such as Mg (magnesium) and Fe (iron), the reaction time between the sample and the reagent is less than 1 minute, and the time until completion is short. On the other hand, since the reaction time of the enzyme reaction depends on the reaction speed of the enzyme and the substrate, it is affected by the concentration of the substrate, temperature, pH, etc., and when the time is long, the reaction time of 2 minutes or more is required. As for the antigen-antibody reaction, the reaction constant between the antigen and the antibody is small, and the reaction is usually not completed even if more than 5 minutes have elapsed after the addition of the antibody. That is to say, the reaction time of enzyme reaction and antigen-antibody reaction is determined by the reaction rate constant of the enzyme itself. In this way, the reaction time between the sample and the reagent varies depending on the item or the concentration of the sample, and there are actually items that do not require a reaction time of even 10 minutes. However, in order to change the photometry time for each item or change the measurement time for each sample, an index reflecting completion is required. Although the reaction process data can be used as the indicator, there is no method for judging the completion of the reaction so far.

Patent Document 1 describes that the concentration of a measurement target substance can be obtained with high accuracy even when a sufficiently long reaction time cannot be obtained. Even if the method described in this document is used, taking into account the error contained in the measurement data, the longer the reaction time is, the smaller the error of the finally obtained concentration of the measurement target substance is. However, there is a problem that it is unclear how much response time should be specifically set. In addition, the optimum reaction time differs depending on the type of the substance to be measured and the reagent used, making it difficult to know the optimum reaction time.

Solution to the problem

The structure of this invention for solving the said subject is as follows.

An automatic analysis device, comprising: a storage mechanism for storing an approximate expression of the time change of the measured value corresponding to each measurement item or each sample; a parameter optimization mechanism for measuring the measured value at each predetermined time , optimizing the parameters of the approximation formula; and judging means, judging whether the variation of the parameter optimized by the parameter optimizing means is within a predetermined range.

The storage mechanism refers to a mechanism for storing information, and may be any mechanism as long as it can store information, such as a semiconductor memory, a hard disk storage device, a Floppy (registered trademark) disk storage device, or a magneto-optical storage device. Most of them are usually installed inside the chassis of the control computer, but they can also be an independent mechanism. The parameter optimization mechanism refers to a mechanism that uses a parameter fitting algorithm such as the least square method to determine each parameter of an approximate expression having a plurality of parameters in a manner that best fits actual data. Usually, it consists of software incorporated into a control computer or a dedicated computer, etc., and hardware for operating the software. It is not limited to this, and any mechanism may be used as long as it can perform parameter fitting to determine parameters.

The judging mechanism is used to capture the form of the parameter calculated by the parameter optimization mechanism asymptotically to a constant value as a parameter change (variation), by comparing with the upper limit value, the lower limit value, or comparing the absolute value of the change amount with Thresholds are compared to determine whether the change is within the specified range, or through multivariate analysis such as Marsh Taguchi, neural network, etc., to determine whether the parameters determined by the parameter optimization mechanism are within the specified range. Usually, it consists of software incorporated into a control computer or a dedicated computer, etc., and hardware for operating the software. The present invention is not limited thereto, and any mechanism may be used as long as the degree of parameter change can be judged.

Preferred embodiments of the present invention are described below.

In the present invention, focusing on the footprints of the sample and reagent in the automatic analyzer from the beginning to the end of the reaction, that is, the reaction process data, the measurement data such as the absorbance of the substance to be measured is obtained successively during the measurement, and an approximate expression of the reaction process is obtained. The concentration of the measurement target substance contained in the sample is predicted by calculating the concentration of the measurement target substance for a certain period of time using the obtained parameter values of the approximation formula.

Solve the above-mentioned problems by calculating an approximate expression using measured values that change over time in the measurement of the object of measurement contained in the sample, and calculating the concentration of the object of measurement for a certain period of time from the obtained approximate expression . In FIG. 3 , the horizontal axis 110 represents the lapse of time, and the vertical axis 120 represents the absorbance. In addition, a dotted line 130 indicates the timing of adding the second reagent, a symbol 140 indicates the actually measured absorbance, and a curve 150 indicates the time change of the absorbance obtained from the approximate formula. In this way, even if the actual reaction time of 10 minutes is not reached, the measured value can be obtained within the approximate photometric time of the point used. By using it in this way, it is not necessary to observe the absorbance until the reaction is completed, and the measured value can be calculated before the reaction is completed.

In addition, since there is a device for comparing the measurement data described by the approximate formula with the sequentially stored time-series data, it is possible to know the optimal response time.

In addition, by calculating the value of the parameter reflecting the state of the measured value that changes with the passage of time from the approximate formula, and storing the calculated parameter value sequentially, the value of the previously stored parameter and the value of the newly stored parameter are stable. By point-calculating the concentration of the measurement target substance for a certain period of time, the concentration can be calculated at the optimum reaction time.

In addition, by calculating the value of the parameter reflecting the state of the measured value that changes with time from the approximate formula, the concentration of the measurement object substance is calculated from the value of the above-mentioned parameter, and the calculated concentration of the measurement object substance is stored. By outputting the concentration of the measurement target substance at the time point when the concentration of the measurement target substance and the newly stored concentration of the measurement target substance stabilize, the concentration can be output at an optimal response time.

In addition, by calculating the value of the parameter reflecting the state of the measured value changing with the passage of time from the approximate formula, the measured value of the substance to be measured is predicted from the value of the parameter, and at a time point where the deviation from the actually obtained measured value is small By calculating the concentration of the substance to be measured, the concentration can be calculated at an optimal reaction time.

In addition, a plurality of formulas including one or more parameters reflecting the state of measured values that change over time are stored, and one of the aforementioned formulas is selected according to the measurement target substance to be measured or the type of reagent to be used. . As for which formula to use, the optimal formula can be determined in advance according to the type of reagent and each item through a pre-verification experiment, or multiple formulas can be used for separate calculations, and residuals from reaction process data obtained over time can be used. The approximate expression in which the difference (the difference between the absorbance obtained by the actual measurement and the absorbance calculated by the approximate expression) decreases is used as the approximate expression for predicting the final concentration value. By calculating the approximation formula, it is possible to approximate the temporal change of absorbance more accurately than before, and it is possible to more easily set an optimum reaction time.

Invention effect

If the present invention can capture the reaction process with high precision, measurement can be performed even if the reaction time is not the current 10 minutes. Therefore, prompt measurement results of urgent samples can be obtained. In addition, not only emergency samples but also ordinary samples can shorten the previous measurement time, and the automatic analyzer of the present invention calculates the reaction using the points immediately after the reaction starts among the multiple points of the total reaction time. The concentration was calculated from the absorbance after the reaction. Since there is no need to observe all the reaction times, the measurement time of the sample can be significantly shortened, and it is expected to improve the efficiency of the biochemical measurement of the automatic analyzer.

In addition, when there is a deviation between the value calculated at the beginning of the reaction and the value calculated at the end of the final reaction, a data alarm can be generated as a reaction abnormality, and the reliability of the data is also improved.

Description of drawings

FIG. 1 is a diagram showing a schematic configuration of an automatic analyzer to which the present invention is applied.

FIG. 2 is a diagram showing a processing flow of the first embodiment.

Fig. 3 is a graph showing the temporal change of absorbance in measurement by an endpoint method.

Fig. 4 is a graph showing the temporal change of absorbance in measurement by a rate method.

FIG. 5 is a graph showing changes in calculated parameter values.

FIG. 6 is a graph showing changes in variances of calculated parameter values.

FIG. 7 is a graph showing the distribution of errors in calculated concentration values.

Fig. 8 is a graph showing reaction process data when TG is measured.

Fig. 9 is a diagram showing the processing flow of the second embodiment.

Fig. 10 is a diagram showing the processing flow of the third embodiment.

FIG. 11 is a diagram showing an example of a table in which optimal approximation expressions and reaction times for combinations of test items and reagents to be used are described.

Fig. 12 is a graph showing the time change of absorbance and the change of absorbance obtained from an approximation formula in the measurement by the endpoint method.

Fig. 13 is a graph showing the time change of absorbance in measurement by the endpoint method and the change of absorbance obtained from an approximation formula.

Fig. 14 is a graph showing the time change of absorbance in measurement by the endpoint method and the change of absorbance obtained from an approximation formula.

Fig. 15 is a diagram showing the processing flow of the fifth embodiment.

Detailed ways

Embodiments of the present invention will be described below using the drawings.

Example

Example 1

Fig. 2 is a diagram showing a schematic configuration of a biochemical automatic analyzer to which the present invention is applied. 1 is the sample tray, 2 is the reagent tray, 3 is the reaction tray, 4 is the reaction tank, 5 is the sampling mechanism, 6 is the pipetting mechanism, 7 is the stirring mechanism, 8 is the photometry mechanism, 9 is the washing mechanism, 10 is the display 11 is an input unit, 12 is a storage unit, 13 is a control unit, 14 is a piezoelectric element driver, 15 is a stirring mechanism controller, 16 is a sample container, 17 and 19 are circular discs, 18 is a reagent bottle, 20 is a refrigerator, 21 is a reaction container, 22 is a reaction container holder, 23 is a driving mechanism, 24 and 27 are probes, 25 and 28 are support shafts, 26 and 29 are arms, 31 is a fixed part, 32 is an electrode, 33 is a nozzle, and 34 is an up and down drive mechanism. The storage unit stores analysis parameters, the number of times that can be analyzed for each reagent bottle, the maximum number of times that can be analyzed, calibration results, analysis results, and the like. The analysis of the sample is performed as follows, and data processing such as sampling, reagent dispensing, stirring, photometry, washing of the reaction vessel, and concentration conversion is sequentially performed.

The sample disk 1 is controlled by the control unit 13 via the display unit 10 . On the sample disk 1, a plurality of sample containers 16 are arranged in a row on the circumference, and move to the bottom of the sampling probe 24 according to the order of the samples to be analyzed. The sample in the sample container 16 is dispensed in a predetermined amount into the reaction container 21 by the sample pump connected to the sample sampling mechanism 5 .

The reaction container 21 into which the sample has been dispensed moves to the first reagent addition position in the reaction tank 4 . In the moved reaction container 16 , a predetermined amount of reagent sucked from the reagent container 18 is added by a reagent pump (not shown) connected to the reagent dispensing probe 6 . The reaction container 21 after the addition of the first reagent is moved to the position of the stirring mechanism 7, and the first stirring is performed. The addition and stirring of the reagents described above are performed for the first to fourth reagents.

The reaction vessel 21 whose contents have been stirred passes through the light beam emitted by the light source, and the absorbance at this time is detected by the light measuring mechanism 8 of the multi-wavelength photometer. The detected absorbance signal enters the control unit 13 and is converted into the concentration of the sample.

The data after density conversion is stored in the storage unit 12 and displayed on the display unit. After the photometry is completed, the reaction container 21 is moved to the position of the washing mechanism 9 to be washed, and then supplied to the next analysis.

Next, the details of the process of converting to the concentration of the sample in the control unit 13 will be described with reference to FIG. 1 . FIG. 1 is a diagram showing a processing procedure of a portion related to concentration conversion in the control unit 13 . First, at the same time as the measurement of a certain inspection item is started for a certain sample, an approximate expression corresponding to the inspection item is selected from among a plurality of approximate expressions representing temporal changes in absorbance in step S5.

As described in the background art, there are roughly two measurement methods: the endpoint method and the rate method, and the change in absorbance is largely different in these two methods. Examples of representative time changes in absorbance by the endpoint method and the rate method are shown in FIGS. 3 and 4 . In FIG. 3 and FIG. 4 , the horizontal axis 110 both represents time, and the vertical axis 120 both represents absorbance. In addition, this reaction is a two-reagent reaction, and after the addition of the second reagent, the change in absorbance for measuring the substance to be measured starts. In FIGS. 3 and 4 , a dotted line 130 indicates the timing at which the second reagent is added. In addition, symbol 140 represents the measured absorbance, and curve 150 represents the temporal change of absorbance obtained from the approximate formula.

In terms of the endpoint method, as shown in Figure 3, the absorbance asymptotically reaches a constant value as the reaction progresses. On the other hand, in the rate method, as shown in Fig. 4, the absorbance changes substantially in a straight line. Therefore, different approximations need to be used in these two methods. In addition, even with the same endpoint method and rate method, time changes are slightly different depending on the item, so it is necessary to prepare multiple expressions and select the best approximate expression corresponding to the item.

For example, the following formula can be prepared and selected as the formula used in the endpoint method. Among them, x is absorbance, t is time, a0, a1, a2, b0, b1, c, d, e, r, s, k1, k2 are parameters.

x＝a0+a1*exp(-k1*t) ...(Number 1)

x＝a0+a1*exp(-k1*t)+a2*exp(-k2*t) ...(Number 2)

x＝c+(1/(b0+b1*t)) ...(Number 3)

x＝d+(e/(exp(r*t)+s)) ...(Number 4)

(Numerical 1) and (Numerical 2) are further generalized to the following formula. Here, n is a natural number, and Σ{} is a symbol representing a sum obtained by changing i in the expression in {} from 1 to n and adding them. It is also possible to set n to a plurality of natural numbers and use (numeral 5).

x＝a0+∑{ai*exp(-ki*t)}                                                                                                                                                                                                                                                   ... (Number 5)

An expression of the following form can be used in the rate method. x is absorbance, t is time, and a, b are parameters. h(t, ψ) is a function containing multiple parameters ψ, t is infinite and asymptotically to zero.

x = A*T+B+H (T, ψ) ... (Number 6)

In the rate method, the absorbance changes linearly with the change of time, so the absorbance x becomes the linear formula x=a*t+b of t in an ideal state, but in the actual reaction, the reaction speed at the initial stage of the reaction is not constant, There is a curve change in the reaction process (lag time). h(t, ψ) in the above formula is a term for accurately approximating the curve portion at the initial stage of the reaction. As a formula that embodies h(t, ψ), for example, the following formula can be used. Among them, x is absorbance, t is time, a, b, c1, d, e, k1, ci, ki, u, v, w, p, q, r are parameters. In addition, n is an arbitrary natural number, and Σ{} is a symbol representing a sum obtained by changing i in the expression in {} from 1 to n and adding them.

x=a*t+b+c1*exp(-k1*t)...(Number 7)

x = a*t+b+∑ {ci*exp (-ki*t)} ... (Number 8)

x＝a*t+b+e/(t+d) ...(Number 9)

x = a*t+b+w/{exp (u*t)+v} ... (number 10)

x = a*t+b+p*log {1+q*exp (r*t)} ... (Number 11)

(Expression 6) to (Expression 11) are expressions for approximating the absorbance change in a linear change after the absorbance changes in a curve with respect to time at the initial stage of the reaction. However, depending on the inspection items, the curve may change again at the end of the reaction. In this case, an expression used for general curve approximation, such as a high-order polynomial, can be used. The expression of such a general curve is expressed in the form shown in (Expression 12) below. Among them, t represents time, x represents absorbance, and φ represents multiple parameters.

x＝g(t, φ) ... (Number 12)

The absorbance is measured a plurality of times over time, but in step S10 , the absorbance data measured once is input from the light measuring mechanism 8 . In the measurement method using light of two wavelengths, light of a wavelength (main wavelength) whose absorbance changes greatly when the color tone changes due to the reaction between the reagent and the sample, and light of a wavelength (sub-wavelength) whose absorbance hardly changes, the main wavelength is input The difference between the absorbance of the light and the absorbance of the sub-wavelength light is taken as absorbance data.

As shown in FIGS. 3 and 4 , in reactions using two or more reagents, a large change in absorbance begins after the addition of the reagent (usually the final reagent) that causes the main absorbance change. Therefore, in step S15, it is determined whether or not the reagent causing a major change in absorbance has been added, and if not added, the process returns to step S10, and the next absorbance data is input. If it has been added, the process moves to S20, and the input absorbance data is stored.

In step S25, it is determined whether or not the number of absorbance data required for calculating the value of the parameter in the formula to minimize the time change of absorbance and the actual time change of absorbance is stored is stored. Usually, in order to calculate the parameter value in the formula, the number of data equal to or greater than the number of parameters is required. When it is determined in step S25 that the required number of data is not stored, the process returns to step S10 and the next absorbance data is input. When the required number of data is stored, the process proceeds to step S30.

In step S30, the parameter value in the formula is calculated so that the formula describing the time change of absorbance and the actual time change of absorbance are as small as possible, and the calculated parameter value is stored in step S31. Specifically, in step S30, in such a manner that the square error between the measured and stored absorbance data and the absorbance at the same time point as the time point at which the absorbance was measured is calculated using (1) to (12), as small as possible, Determine the parameter values in the formula. For the calculation of parameter values, the conventional least square calculation method can be used, but as a method that can handle various forms of calculation expressions, for example, the steepest descent method is used to calculate the parameter value with the smallest square error.

In step S40, it is determined whether or not parameters are stored enough times for density calculation. In subsequent calculations, the concentration value is calculated from the parameters, but generally, when the number of observed data is small, errors included in the calculated concentration value increase. Therefore, in this embodiment, in order to prevent the output of concentration values containing many errors, it is necessary to determine the minimum number of calculations for the parameters required for concentration calculation. In step S40, it is checked whether the calculation of parameters more than this number of times has been performed. . If the calculation of parameters for the necessary number of times has not been performed, the process returns to S10 and the next absorbance data is input. When the parameters more than the required number of times have been calculated, the process proceeds to step S45.

In step S45, the magnitude of the temporal variation of the calculated parameter is calculated. In the present invention, the process of measuring the absorbance and obtaining the parameters of the formula is repeated after the start of the reaction. The calculation to obtain the parameters of the formula is a process of estimating the parameters of the formula so as to match the observed absorbance as much as possible. However, in the early stage of the reaction, when the number of absorbance data is small, due to the error contained in the data, the estimated parameters may not be accurate. The included error also increases. As the absorbance data increases over time, the random errors contained in the absorbance data cancel out and the errors in the inferred parameters decrease. Therefore, in the initial stage of the reaction, the value of the estimated parameter also fluctuates every time due to the influence of the error contained in the absorbance, but as the number of data increases, the variation of the parameter decreases and converges to the optimal value. The parameter value at each time point of the observed absorbance was obtained from the actual absorbance data, and an example after plotting is shown in FIG. 5 . The horizontal axis 210 represents time, and the vertical axis 220 represents the value of the parameter. Reference numeral 240 represents the value of the parameter calculated at each time.

In step S45, the time variation of the parameter is digitized. As a method of quantifying the change of the parameter, various methods can be used. For example, the difference from the value of the parameter calculated last time, the variance of the parameter several times before, or the difference between the maximum value and the minimum value of the parameter several times can be used. Poor wait. As the time variation of the parameter, at a certain point in time, the variance of the parameter values of five times up to the fourth time in total was obtained from the time point, and an example after plotting is shown in FIG. 6 . The horizontal axis 110 represents the passage of time, and the vertical axis 320 represents the variance of the parameter values. Symbol 340 represents the value of the variance calculated at each time.

In step S50, the time variation of the parameter obtained in step S40 is compared with a predetermined threshold value. Here, when the parameter change is equal to or less than a predetermined threshold, it is determined that sufficient absorbance data for calculating the concentration of the substance to be measured has been accumulated, and therefore the processing proceeds to step S65 to calculate the concentration. If the parameter change is greater than the predetermined threshold value, it is considered that sufficient absorbance data for calculating the concentration value has not been accumulated, so the process moves to step S55 to further check whether there is next data. The threshold for comparing parameter fluctuations is set in advance according to the purpose of the device so as to obtain the required measurement accuracy. However, the user can change it according to the purpose of inspection. In addition, it is also possible to set different values for each inspection item. For example, in the example shown in FIG. 6 , when the threshold is set to 50, a dotted line 330 indicates the threshold. In this case, since the parameter variation was lower than the threshold value at time 35, it was judged that sufficient absorbance for calculating the concentration value was accumulated at this time point.

When there are a plurality of parameters, a threshold value is set for the variation of all parameters, and when the variation of all parameters is lower than the threshold value, the process proceeds to S65. However, regarding this determination condition, various examples may be taken into consideration, and when the variation of some selected parameters among a plurality of parameters is lower than the threshold value, the process may be shifted to step S65.

When it is determined in step S55 that there is next data, the process returns to step S10 and the next absorbance data is input. If there is no next absorbance data, it is judged that parameters with sufficient accuracy cannot be obtained even after a predetermined reaction time elapses, and thus abnormal data is recorded in step S60.

In step S65, the concentration of the measurement target substance is calculated using the parameters calculated in step S30. In the case of the endpoint method, in order to calculate the concentration, the absorbance at the time point at which the absorbance does not change after a sufficient time has elapsed is converted into the concentration. In the present invention, the parameter values calculated in step S30 are substituted into the approximate formula selected in step S5, and the value of the formula when the time is changed to infinity is used as the absorbance after a sufficient time has elapsed. Specifically, a0 in (Expression 1), (Expression 2), and (Expression 5), c in (Expression 3), and d in (Expression 4) are the absorbance to be obtained. According to the present invention, even if the absorbance itself changes, the concentration can be calculated if the parameter is a constant value, so it is possible to perform high-precision measurement with shorter reaction time than conventional ones. As a method for converting the absorbance obtained from the above parameters into the concentration of the substance to be measured, for example, a conventional method using a calibration curve can be used.

In the case of the rate method, the parameter value calculated in step S30 is substituted into the approximation formula selected in step S5, the slope of the straight line is calculated, and the obtained slope is converted into the concentration value of the measurement object substance. Specifically, in (Expression 6) to (Expression 11), the value of the parameter value a corresponds to the slope of the straight line portion. In the formula of a general curve such as (Expression 12), the portion where the slope changes the least is regarded as a straight line. That is, the quadratic differential of time g"(t, φ) is obtained, and the time point ta at which the absolute value of g"(t, φ) becomes the minimum is regarded as the linear portion. Let the first time differential g'(ta, φ) of ta be the slope of the straight line. In the rate method, depending on the conditions, the length and shape of the curve portion at the initial stage of the reaction are different. In the past, it was difficult to determine the straight line portion, and it was difficult to determine the optimum reaction time for judging the straight line portion. However, according to the present invention, the curve of the rate method can be easily determined. The slope of the straight line, the reaction time can also be optimized. As a method of converting the slope of the straight line into the concentration of the measurement target substance, for example, a conventional method using a calibration curve can be used.

In step S70, an error with respect to the obtained density value is calculated. As described in the description of step S45, when the number of absorbance data is small at the initial stage of the reaction, errors contained in the estimated parameters also increase due to errors contained in the data. As the absorbance data increases over time, the random errors contained in the absorbance data cancel out and the errors in the inferred parameters decrease. Therefore, when there are few absorbance data, the error contained in the final converted concentration value is large, and the larger the number of data, the smaller the error.

The distribution of errors in the concentration values at intermediate time points relative to the concentration values calculated using all the absorbance data is shown, for example, in FIG. 7 . FIG. 7 is a diagram schematically showing the results of measuring 20 times a quality control substance whose concentration is known and examining the error in the concentration value calculated at each time point. In FIG. 7, the horizontal axis 110 represents the passage of time, and the vertical axis 420 represents the error. Symbol 440 represents the average value of the error distribution at each time point, and line segment 460 represents the standard deviation of the error at each time point. As time goes by, the mean value of the error and the standard deviation of the error decrease.

The error distribution of the concentration value calculated at each time point is studied in advance using a plurality of data, and the relationship between the time point and the error distribution is stored as a table. For example, the average value and standard deviation of errors at each time point are stored in a table. In step S70, the error at the point in time at which the concentration value was calculated is calculated from the stored table of the relationship between the point in time and the error. For example, by displaying the average value and variance value of the error, the user can know the error range of the measurement result. In addition, by displaying the error distribution at each time point accumulated in the table, it becomes reference information when setting the minimum number of parameter calculations used in step S40, the threshold value of parameter variation used in step S50, and the like.

In the above-mentioned first embodiment, the parameters included in the formula are obtained from the absorbance data multiple times during the reaction time, and the time variation of the parameters is used to determine whether the time required for calculating the concentration has elapsed. Therefore, even when it is unclear how much response time should be set specifically, the response time can be automatically determined. In addition, even if the optimum reaction time differs depending on the type of the substance to be measured and the reagent used, the reaction time can be determined automatically.

In addition, in order to estimate the error corresponding to the number of time-series data used, that is, the response time, the device user can quantitatively know how much the response time is set and how much error will be obtained. It is also possible to set the optimal reaction time according to the project or purpose.

A specific method of calculating the concentration of the rapid measurement is shown by taking the item of TG (neutral fat) as an example.

As for the measurement method of TG, the endpoint method of the 2-reagent method was used. The reaction process is shown in FIG. 8. After the addition of the second reagent, the absorbance increases, and when the reaction proceeds for a certain period of time, the absorbance becomes substantially constant. The formula that can be approximated with high precision in this reaction process is preferably selected and set when studying or measuring the calibrator in advance, but it is possible to use a plurality of formulas (Expression 1) to (Expression 5) to parallelize each other. In the determination part of the concentration calculation, it is judged which approximate formula to use for concentration calculation based on the magnitude of the residual (the difference between the absorbance value calculated by the approximate formula and the absorbance value obtained by actual measurement).

In Example 1, the approximation formula used for the rapid calculation of TG when the reagent R is used is set in advance to (Expression 1). In the automatic analyzer having the device configuration shown in FIG. 2, the sample and the reagent are added to the reaction vessel shown by 21 in FIG. When P1 is the first photometric point at which the rise starts, for example, in this analyzer, absorbance is obtained at P1 to P18 over time ( FIG. 8 ). Since the calculation of the approximate formula requires at least two or more absorbance values, the photometry points required for the concentration calculation are calculated from the time point when P2 is measured.

Approximate calculations are performed from the points of P1 and P2, approximate calculations are performed from the points of P1 to P3, and calculations of approximate formulas are performed from P2 to P4, P1 to P5... every time the absorbance is measured.

The parameters calculated by approximate calculation, such as the value of the final absorbance value A(a0) calculated by the approximate formula, are shown in FIG. 5 . As the photometering points increase, the approximation accuracy improves and the value gradually stabilizes. The parameter k other than A and the absorbance x(t) at an arbitrary time t may also be used as the value for determination. In addition, as a criterion for evaluating the stability of the parameter value, for example, as shown in FIG. 6 , the variance of the value of the calculated parameter quintuple (P2 to P6) is obtained. If the parameter is stable, the value of its variance also decreases, but if, for example, a threshold value is set for the value of the variance and the value is below the threshold value, the parameter is judged to be stable, and the measured value is predicted and calculated using the approximate expression and parameter value at that point in time. . Here, as shown in FIG. 6 , when the value of the variance is 10 or less, it is determined that the parameters are stable, and the density is calculated from the approximate expression of the photometric points P1 to P15 (photometric point 35 ).

The concentration calculation method used by the automatic analyzer is generally obtained from the following formula.

Cx＝{k×(absorbance of sample-absorbance of standard solution 1)}×device constant ...(Number 13)

The k in the formula is the k factor, which can be obtained from the calibration results. The concentration Cx of the measurement target substance to be obtained can be obtained from the absorbance a0 at an arbitrary time or at a point in time when the reaction reaches an equilibrium state. Alternatively, the absorbance Ct at the usual measurement completion time point may be calculated from an approximate formula and output as the predicted value Cm.

Example 2

The biochemical automatic analyzer of the second embodiment of the present invention is also the same as that of the first embodiment, and its configuration is schematically shown in FIG. 2 . Since operations other than the control unit 13 are the same as those in the first embodiment, detailed description thereof will be omitted.

Details of the process of converting the absorbance of the control unit into the concentration of the sample in the second embodiment will be described with reference to FIG. 9 . In addition, since the processing denoted by the same code|symbol as FIG. 1 is the same as the process denoted by the same code|symbol in FIG. 1, detailed description is abbreviate|omitted below.

From the start of the process, the processes of step S5, step S10, step S15, step S20, step S30, and step S35 are the same as those of the first embodiment shown in FIG. 1 . After the parameters are calculated in step S35, in this embodiment, the concentration value of the measurement target substance is calculated using the calculated parameters in step S65. The details of the processing of calculating the density value from the parameters are the same as the processing of step S65 described in the first embodiment. The calculated density value is stored in step S100.

In step S110, it is determined whether or not enough concentration values have been calculated and stored for the calculated concentration value to be the final measurement result. As described in the description of step S40 in the first embodiment, if the number of observed data is usually small, the error included in the calculated concentration value increases. Therefore, in this embodiment, in order to prevent the output of the concentration value containing many errors, the minimum number of calculations required to calculate the concentration that will become the final measurement result is determined, and it is checked in step S110 whether the concentration value has been calculated more than the number of times. calculated. If the calculation of the required number of concentration values has not been performed, the process returns to step S10, and the next absorbance data is input. When the density value equal to or greater than the required count has been calculated, the process proceeds to step S120.

In step S115, the magnitude of the temporal variation of the calculated density value is calculated. In the present invention, the process of measuring the absorbance, obtaining the parameters of the formula, and calculating the concentration value is repeated after the start of the reaction. The calculation to obtain the parameters of the formula is to estimate the parameters of the formula so that the parameters of the formula are as close as possible to the observed absorbance. However, in the early stage of the reaction, when the number of absorbance data is small, due to the error contained in the data, the estimated parameters may not be accurate. The error included also increases, and the error included in the concentration value calculated from the result also increases. As time goes by and the absorbance data increases, the random error contained in the absorbance data cancels out, the error of the inferred parameters also decreases, and the error contained in the calculated concentration value also decreases. Therefore, at the initial stage of the reaction, due to the influence of the error contained in the absorbance, the concentration value calculated every time also fluctuates. As the number of data increases, the fluctuation of the concentration value decreases and converges to the optimal value. Concerning the concentration value, the same time variation as the parameter value shown in FIG. 5 is shown.

In step S115, the magnitude of the time variation of the above-mentioned concentration value is converted into a numerical value. Various methods can be used to quantify the time variation of the concentration value. For example, the difference from the previous concentration value, the variance of multiple density values, or the difference between the maximum value and the minimum value of multiple density values can be used. Poor wait. As the time variation of the concentration value, at a certain point in time, when the variance of the concentration value of five times up to four times in total is obtained from the time point, the same change as the parameter variation shown in FIG. 6 is shown.

In step S120, the time variation of the concentration value obtained in step S115 is compared with a predetermined threshold value. Here, when the temporal variation of the concentration value is equal to or less than a predetermined threshold, it is determined that sufficient absorbance data for calculating the concentration of the measurement target substance has been accumulated, and therefore the processing proceeds to step S70 to calculate an error. If the time variation of the concentration value is greater than the predetermined threshold value, it is considered that sufficient absorbance data for calculating the concentration value has not been accumulated, so the process moves to step S55 to further check whether there is the next data. The threshold value for comparing fluctuations in concentration values is set in advance according to the purpose of the device so that the required measurement accuracy can be obtained. However, the user can change it according to the purpose of inspection. In addition, it is also possible to set different values for each inspection item.

The processing of steps S55, S60, and S70 is the same as the processing of the same symbols in the first embodiment, and thus description thereof will be omitted.

In the above-mentioned second embodiment, the parameters contained in the calculation formula are obtained from the absorbance data multiple times during the reaction time, the concentration value is calculated, and the time fluctuation of the concentration value is used to determine whether the time required for calculating the concentration has passed. time. Therefore, even when it is unclear how much measurement time should be set specifically, the reaction time can be automatically determined. In addition, even if the optimum reaction time differs depending on the type of the substance to be measured and the reagent used, the reaction time can be automatically determined.

In addition, in order to estimate the error corresponding to the number of time-series data used, that is, the response time, the device user can quantitatively know how much the response time is set and how much error will be obtained. It is also possible to set the optimal reaction time according to the project or purpose.

Example 3

The biochemical automatic analyzer of the third embodiment of the present invention is also the same as that of the first embodiment, and its configuration is schematically shown in FIG. 2 . Since operations other than the control unit 13 are the same as those in the first embodiment, detailed description thereof will be omitted.

Details of the process of converting the absorbance of the control unit into the concentration of the sample in the third embodiment will be described with reference to FIG. 10 . In addition, since the processing denoted by the same code|symbol as FIG. 1 is the same as the process denoted by the same code|symbol in FIG. 1, detailed description is abbreviate|omitted below.

First, an approximate formula is selected in step S5, and a reaction time is selected in step S210. The control unit 13 stores a table 500 , as shown in FIG. 11 , in which optimal approximation expressions and reaction times for combinations of test items (measurement target substances) and reagents to be used are described. In the column 510, inspection items are described, and in the column 520, the types of reagents are described. Inspection items indicate measurement target substances. Column 530 describes the type of the optimal approximation formula for the test item and the type of reagent, and column 540 describes the optimal reaction time. Considering the combination of inspection items and reagents, the table 500 is used to select the optimal approximate formula in step S5, and the table 500 is also used to select the optimal reaction time in step S210. In addition, the content of the table may be configured to be changeable by the user.

In step S10, absorbance data is input from the photometric mechanism 8, and in step S20, the absorbance data is stored. In step S25, it is determined whether or not the reaction time selected in step S210 has elapsed, and if not elapsed, the process returns to step S10, and the next absorbance data is input. If it has expired, move the process to step S30.

In step S30, the parameters of the formula selected in step S5 are calculated using the stored absorbance data. Furthermore, in step S65, the parameter value calculated in step S30 is converted into the concentration of the chemical component of a measuring object. In step S70, an error corresponding to the reaction time is calculated.

In the above-mentioned third embodiment, a plurality of formulas including one or more parameters describing the time change of absorbance are stored, and by selecting the optimum formula according to the combination of the substance to be measured and the reagent, the formula can be compared with the conventional method. It is easier to set the optimal reaction time than expressing the time change of absorbance with high accuracy. For example, Figure 12 uses the absorbance data shown in Figure 8 of a certain inspection item TG (neutral fat) measured by the end point method to obtain the parameter value of (number 1), and substitute the obtained parameter value into (number 1), and An example of plotting the obtained absorbance change curve (reaction process curve) and actual absorbance data on the same graph. The horizontal axis 110 represents the passage of time, and the vertical axis 120 represents the absorbance. In addition, a symbol 140 represents the absorbance actually measured at each time point, and a curve 150 represents the time change of the absorbance calculated by the approximate formula. In this example, the actually obtained time change in absorbance agrees well with the time change represented by (Numerical 1).

FIG. 13 shows the results of similar processing using the absorbance data of another test item TP (total protein). As can be clearly seen from the figure, the error between the actually measured absorbance and the absorbance calculated by the approximate formula is large after a sufficient time has elapsed. It can be seen from this example that (Number 1) is not suitable for expressing the time change of the absorbance of this inspection item. The result of processing the absorbance data using (2) is shown in FIG. 14 . As can be seen from this figure, (Expression 2) is suitable for expressing the time change of the absorbance data of the test item.

In addition, in the rate method, the change in absorbance becomes the slope of the linear portion to obtain the activity value of the enzyme as the measurement target substance, but when (equation 1) is used, it is difficult to clearly detect the linear portion. In items measured by the velocity method, by using (Expression 7) to (Expression 11), the slope of the straight line portion can be easily detected as the value of the parameter a. When (Expression 12) is used, the slope of the straight line can be easily calculated as the time primary differential at the point where the secondary differential of time becomes the minimum.

As described above, it is not possible to express the temporal change of absorbance with sufficient accuracy for various test items and combinations of reagents using a single formula. As in the present embodiment, by selecting and using a plurality of calculation formulas, time changes can be expressed with high precision for each test item and reagent combination, and high precision results can be obtained within a short reaction time.

Example 4

In the fourth embodiment, the device configuration shown in FIG. 2 and the processing steps shown in FIG. 1 are the same as those in the first embodiment. Since only the approximation formula selected in step S5 of FIG. 1 is different from the approximation parameter calculation method in step S30 and the density calculation in step S65, these three processing steps will be described in detail.

In the first embodiment, a formula representing absorbance x as a function of time t was used as a formula that can be selected in step S5, but in this embodiment, a differential equation is used as the formula. In order to theoretically explain the temporal change of absorbance, differential equations are often used, but in this example, the theoretical equations can be directly used. For example, let time be t, absorbance be x, Σ{} be the symbol representing the sum obtained by changing i in the expression in {} from 0 to n and adding, and n be 1 or more Integers, if fi(t, x) is set as a function including t or x or any number of times of time differentiation of x, including the case where fi(t, x) is a constant, when qi is set as a parameter, you can use A differential equation of the form expressed by the following expression.

∑{qi*fi(t, x)}＝0                                                                                                                                                                                                                                                 back time ... (Number 13)

In addition, as a special case of (Expression 13), the differential equation shown in (Expression 14) can be used. Wherein, the n-time time differential of absorbance x at time t is set as x[n](t), and p and pi are set as parameters.

p+∑{pi*x[n](t)}=0

More specifically, for example, the following differential equation can be used. Among them, x(t)^2 means the square of x(t).

p+p0*x (t)+p1*x [1] (t) = 0 ... (number 15)

p+p0*x(t)+p1*x[1](t)+p2*x[2](t)＝0                                                                        ...

q2*x (t)^2+q3*x [1] (t) = 0 ... (Number 17)

q1*x(t)+q2*x(t)^2+q3*x[1](t)＝0                                                                                                      ... (Number 18)

q0+q1*x(t)+q2*x(t)^2+q3*x[1](t)＝0                                                                                                                                                                   ... (Number 19)

In step S30, the values of the parameters included in (13) and (14) are determined using the stored absorbance data. Since the absorbance is stored as time-series data, the time differential can be approximately calculated by calculating the difference. Therefore, in order to obtain the values of fi(t,x) of (Expression 13) and x[n](t) of (Expression 14) corresponding to the time t at which the absorbance is measured, these values are obtained at multiple time points , (Number 13) and (Number 14) are respectively represented by the linear combination of fi(t, x) and x[n](t), so the parameters p, pi, and qi can be easily obtained by using the least square method value. Here, as an example, a case where the time change of the absorbance x is expressed by the formula shown in (Expression 15) will be described. In addition, the m+1 absorbance is measured to obtain the absorbance of x0 to xm. (Equation 15) can be transformed into the following form by setting x(t) to the left and setting the remaining terms to the right.

x (t) = r1*x [1] (t)+r ... (number 20)

In this case, y1 is obtained by calculations such as y1=(x2-x0)/(2*h) and y2=(x3-x1)/(2*h) as an amount corresponding to the primary time differential. m-1 difference values in ~y(m-1). In (Number 20), when xi and yi are substituted for x(t) and x[1](t), (Number 20) is represented by (Number 21). Wherein, i=1~m-1.

xi=p1*yi+p ...(Number 21)

Since the relationship represented by (number 20) is not completely consistent with the observed absorbance, the value on the right side of (number 21) is inconsistent. Therefore, the parameters r1 and r are determined by the method of least squares in such a way that the difference between the right side and the left side is minimized. Here, when xi is a vector X arranged vertically, and A is a matrix of m-1 rows and 2 columns shown below, and R=(r1, r)', the relationship of (21) can be expressed by (22) express. Among them, the symbol ' means transposition.

y1 1

y2 1

y3 1

: :

y(m-1) 1

X＝AR ...(Number 22)

If the characteristic equation of (Number 22) is to be solved, the least square solution is obtained by (Number 23). where inv() represents the inverse matrix of the matrix in ().

R = {inv (a′a)} a′x ... (Number 23)

When expressing the relationship between absorbance and time, expressing it using a differential equation reduces the number of parameters to be obtained compared to expressing it as a function of t normally. Also, when the differential equation is represented by a linear combination of functions of the measured absorbance data like (Expression 13) and (Expression 14), the parameters can be easily calculated by the least square method as described above.

In step S65, the concentration of the component to be measured is calculated using the parameter values obtained in step S30. In the endpoint method, the absorbance becomes a constant value when a sufficient time has elapsed. That is, since there is no time change, the time change is 0. Therefore, in (Equation 13) and (Equation 14), from the value of x(t) when all x[n](t) of n≥1 are 0, it can be obtained that the absorbance becomes constant after a sufficient time has elapsed. absorbance at . For example, in (Number 15) and (Number 16), since x=-p/p0 when x[1](t)=0 and x[2](t)=0, this value is taken as of absorbance. The concentration of the substance to be measured is converted from the absorbance using a calibration curve or the like.

In the rate method, when a sufficient time has elapsed, the absorbance changes linearly with respect to time, and the concentration of the component to be measured is calculated from the slope of the straight line. Therefore, the value of x[1](t) when all x[n](t) of n≧2 are 0 after a sufficient time has passed can be used as the slope of the time change in absorbance when a sufficient time has passed. The concentration value of the substance to be measured is converted from the slope, using a calibration curve, or the like.

As described above, in the fourth embodiment, by using the formula expressing the time change of absorbance as a differential equation, the differential equation derived from the chemical reaction velocity theory can be used directly. Compared to this, the number of parameters is reduced and the calculation of the least squares method for determining the parameters becomes easier.

Example 5

The automatic biochemical analysis apparatus of the fifth embodiment of the present invention is also the same as the first embodiment, and the outline of its structure is shown in FIG.

Details of the process of converting the absorbance in the control unit into the concentration of the sample in the fifth embodiment will be described with reference to FIG. 15 . In addition, since the processing denoted by the same code|symbol as FIG. 1 is the same as the process denoted by the same code|symbol in FIG. 1, detailed description is abbreviate|omitted below.

From the start of the process, the processes of step S5, step S10, step S15, step S20, step S30, and step S35 are the same as those of the first embodiment shown in FIG. 1 . After the parameters are calculated in step S35, in this embodiment, through step S200, the calculated parameters are substituted into the approximate formula, and the predicted value of the absorbance at the current time point is calculated using the approximate formula.

In step S210, the error between the current predicted value of absorbance obtained in step S200 and the actually measured absorbance input in step S10 is calculated.

In step S220, the error in the absorbance obtained in step S210 is compared with a predetermined threshold value. Here, when the time variation of the concentration value is equal to or less than a predetermined threshold, it is determined that sufficient absorbance data for calculating the concentration of the measurement target substance has been accumulated, and therefore the processing proceeds to step S70 to calculate an error. If the time variation of the concentration value is greater than the predetermined threshold value, it is considered that sufficient absorbance data for calculating the concentration value has not been accumulated, so the process moves to step S55 to further check whether there is the next data. The threshold value for comparing the absorbance error is set in advance according to the purpose of the device so that the required measurement accuracy can be obtained. However, the user can change it according to the purpose of inspection. In addition, it is also possible to set different values for each inspection item.

The processing of steps S55, S60, and S70 is the same as the processing of the same symbols in the first embodiment, and thus description thereof will be omitted.

In the above-mentioned fifth embodiment, the parameters included in the approximate formula are obtained from the absorbance data in the reaction time, and the predicted value of the absorbance at the time point when the absorbance is measured is obtained from the approximate formula. Further, the error between the predicted value of absorbance and the actually measured absorbance was calculated. The longer the reaction time and the more measured absorbance, the higher the accuracy of the approximation and the smaller the error. Therefore, it can be determined whether or not the time required to calculate the concentration has elapsed based on the magnitude of the error. It is also possible to automatically determine the reaction time when it is not clear how much measurement time should be set specifically. In addition, the optimum reaction time varies depending on the type of the substance to be measured and the reagent used, and the reaction time can also be determined automatically.

Symbol Description

1 sample tray

2 reagent trays

3 reaction trays

4 reaction tanks

5 sampling mechanism

6 pipetting mechanism

7 Stirring mechanism

8 Light Metering Mechanism

9 washing mechanism

10 Display

11 Input section

12 storage department

13 Control Department

14 Piezo element driver

15 Stirring Mechanism Controller

16 sample container

17, 19 round plate

18 reagent bottles

20 cold storage

21 reaction vessel

22 Reaction Vessel Holder

23 drive mechanism

24, 27 probe

25, 28 Support shaft

26, 29 arms

31 fixed part

32 electrodes

33 nozzles

34 Up and down driving mechanism

110 represents the time-lapse axis

120 Axis representing absorbance

130 Dashed line representing the moment of addition of the reagent causing the main reaction

140 Symbol for measured absorbance

150 Temporal change in absorbance calculated using approximate formula

220 represents the axis of the parameter value

240 A symbol representing the parameter value calculated at each time

320 represents the axis of the variance of the parameter

330 is the dashed line representing the threshold set for the variance

340 The sign indicating the variance of the parameter calculated at each time

420 An axis representing the error of the concentration value

440 The sign indicating the average value of the error of the concentration value

460 A line segment representing the standard deviation of the error of the concentration value

500 A table that records the optimal approximate formula and reaction time for the combination of inspection items and reagents used

510 List of inspection items

520 A column describing the type of reagent

530 List of types of approximation expressions

540 is a column that records the reaction time.

Claims (22)

1. an automatic analysing apparatus, is characterized in that, possesses:

Storing mechanism, stores the approximate expression changing with each mensuration project or each sample time corresponding, measured value;

Parameter optimization mechanism, when the mensuration of the measured value of each stipulated time, by the parameter optimization of described approximate expression; With

Decision mechanism, judges that the variation of having carried out optimized parameter by described parameter optimization mechanism is whether in the scope of predetermining.

2. automatic analysing apparatus as claimed in claim 1, is characterized in that,

Possess measured value and calculate mechanism, in described decision mechanism, be judged to be the time point in the scope of predetermining, based on having carried out optimized parameter by described parameter optimization mechanism, determine described approximate expression, utilize determined approximate expression to calculate the measured value at reaction deadline point.

3. automatic analysing apparatus as claimed in claim 1, is characterized in that,

The scope that the storage of described decision mechanism is predetermined is as reference space, and mahalanobis distance is calculated in the variation of the parameter based on current point in time, determines whether in the scope of predetermining.

4. automatic analysing apparatus as claimed in claim 1, is characterized in that,

Described decision mechanism determines whether in the scope of predetermining by neural network.

5. an automatic analysing apparatus, is characterized in that, possesses:

Storing mechanism, stores the approximate expression changing with each mensuration project or each sample time corresponding, measured value;

Parameter optimization mechanism, when the mensuration of the measured value of each stipulated time, by the parameter optimization of described approximate expression;

Determination object material concentration is calculated mechanism, based on having carried out by described parameter optimization mechanism the concentration that optimized parameter is calculated determination object material;

Decision mechanism, judges by this determination object material concentration and calculates the concentration of determination object material that mechanism calculates and the difference of the concentration of the determination object material of having stored whether in the scope of predetermining; With

Output mechanism, Gai decision mechanism is judged to be the concentration of the time point output determination object material in the scope of predetermining.

6. an automatic analysing apparatus, is characterized in that, possesses:

Storing mechanism, stores the approximate expression changing with each mensuration project or each sample time corresponding, measured value;

Parameter optimization mechanism, when the mensuration of the measured value of each stipulated time, by the parameter optimization of described approximate expression;

Determination object material concentration projecting body, based on having carried out by described parameter optimization mechanism the concentration that optimized parameter is predicted determination object material;

Decision mechanism, judges that the concentration of the determination object material calculated by this determination object material concentration projecting body and the difference of the concentration of the determination object material of having stored are whether in the scope of predetermining; With

Output mechanism, Gai decision mechanism is judged to be the concentration of the time point output determination object material in the scope of predetermining.

7. an automatic analysing apparatus, is characterized in that, possesses:

Storing mechanism, stores the approximate expression changing with each mensuration project or each sample time corresponding, measured value;

Parameter optimization mechanism, when the mensuration of measured value sample, each stipulated time of concentration known in advance, by the parameter optimization of described approximate expression;

Measure material concentration and calculate mechanism, based on having carried out optimized parameter by described parameter optimization mechanism, calculate the concentration of measuring material;

Decision mechanism, judge by this mensurations material concentration calculate the concentration of the mensuration material that mechanism calculates and in advance the difference of known concentration whether in the scope of predetermining; With

Time point storing mechanism, stores this decision mechanism and is judged to be the time point in the scope of predetermining.

8. automatic analysing apparatus as claimed in claim 7, is characterized in that,

Time point in being stored in described time point storing mechanism, based on having carried out optimized parameter by described parameter optimization mechanism, calculates by described mensuration material concentration the concentration that general sample is calculated by mechanism.

9. automatic analysing apparatus as claimed in claim 1, is characterized in that,

Possess approximate expression selection mechanism, described storing mechanism stores a plurality of described approximate expressions, and described approximate expression selection mechanism, according to the kind of the determination object material as measuring object or the reagent that uses, is selected a kind of formula from described approximate expression.

10. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a0+a1*exp(-k1*t)+a2*exp(-k2*t)，

Described parameter is a0, a1, a2, k1, k2.

11. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, ∑ { } is made as the i that represents to make the formula { } in from 1 be changed to n and be added and and symbol, by n be made as natural number, when the symbol of expression multiplication is made as to *, at least one of described approximate expression is following formula:

x=a0+∑{ai*exp(-ki*t)｝，

Described parameter is a0, ai, ki.

12. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=c+(1/(b0+b1*t))，

Described parameter is b0, b1, c.

13. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as X, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=d+(e/(exp(r*t)+s))，

Described parameter is d, e, r, s.

14. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, by ψ be made as a plurality of parameters, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a*t+b+h(t，ψ)，

Described parameter is a, b, ψ, is calculated the concentration of determination object material by the value of described a,

Described h (t, ψ) is for being similar to accurately the item of the curved portion of initial reaction stage.

15. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a*t+b+c1*exp(-k1*t)，

Described parameter is a, b, c1, k1.

16. automatic analysing apparatus as claimed in claim 1, is characterized in that,

At least one of described approximate expression by measure be constantly made as t, by calculated value be made as x, ∑ { } is made as the i that represents to make the formula { } in from 1 be changed to n and be added and and symbol, by n be made as natural number, when the symbol of expression multiplication is made as to *, its formula is following formula:

x=a*t+b+∑{ci*exp(-ki*t)｝，

Described parameter is a, b, ci, ki.

17. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by measured value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a*t+b+e/(t+d)，

Described parameter is a, b, d, e.

18. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by measured value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a*t+b+w/{exp(u*t)+v｝，

Described parameter is a, b, u, v, w.

19. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by calculated value be made as x, when the symbol that represents multiplication is made as to *, at least one of described approximate expression is following formula:

x=a*t+b+p*log{1+q*exp(r*t)｝，

Described parameter is a, b, p, q, r.

20. automatic analysing apparatus as claimed in claim 1, is characterized in that,

Mensuration is constantly made as t, by calculated value be made as x, when φ is made as to a plurality of parameter, at least one of described approximate expression is following formula:

x=g(t，φ)，

Described parameter is φ, by the time second differential g of described formula, and " value of the time one subdifferential g ' (t, φ) of the described formula when absolute value of (t, φ) becomes minimum t is calculated the concentration of determination object material.

21. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measuring constantly, be made as t, absorbance be made as to x, n the time diffusion of the moment t of measured value is made as to x[n] (t), ∑ { } is made as the i that represents to make the formula { } in from 0 be changed to n and be added and obtain and symbol, by n be made as more than 1 integer, when the symbol of expression multiplication is made as to *, described approximate expression is following formula:

p+∑{pi*x[n](t)｝=0，

Described parameter is p0, pi.

22. automatic analysing apparatus as claimed in claim 1, is characterized in that,

By measure be constantly made as t, by measured value be made as x, ∑ { } is made as the i that represents to make the formula { } in from 0 be changed to n and be added and obtain and symbol, by n be made as more than 1 integer, by fi (t, x) be made as the time diffusion of the arbitrary number of times that comprises t or x or x function, the symbol that represents multiplication is made as to *, also comprises fi (t, while x) being the situation of constant, described approximate expression is following formula:

∑{qi*fi(t，x)｝=0，

Described parameter is qi.

Images